homelab-codex-ws/services/ha-diag-agent/tests/test_config.py
oskar f2ba81bcc1 fix(ha-diag): node_name from env + fail-fast on unknown — evt-unknown-* files poisoned observer checkpoint (see d5139c9)
An "unknown"/empty node_name silently produced evt-unknown-* event files. One
landed in events/piha/ and, being lexically greater than every evt-piha-* name,
pinned the observer's (then lexical) checkpoint and blocked PIHA for ~34 days.

Defense in depth, two layers, both refuse rather than emit poison:
- config.py: node_name field_validator rejects ""/"unknown" (case-insensitive,
  trimmed); validate_default=True so the "unknown" default itself is rejected
  when NODE_NAME never reaches the process. main() catches ValidationError and
  exits 1 with a clear FATAL message instead of a raw traceback.
- event_emitter.py: EventEmitter.__init__ guards node_name at the exact poison
  site (node_name is embedded in the filename), so no future call path can
  reintroduce evt-unknown-*.

Precedence unchanged and correct: Settings.load() uses os.environ.setdefault, so
env NODE_NAME wins over YAML; YAML only supplies node_name when env is absent.
api.py never writes event files (only /health, /trigger), so its _node_name
global cannot produce poison — verified, left as-is.

Tests (services/ha-diag-agent/tests/): new test_config.py (env load, env>YAML
precedence, fail-fast on unknown/empty/whitespace/default/YAML-unknown) and
EventEmitter guard tests. Full unit suite: 139 passed, 0 regressions.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-07-15 19:48:58 +02:00

142 lines
4.7 KiB
Python

"""Tests for Settings / node_name fail-fast.
Regression coverage for the evt-unknown-* poison bug: a node_name of "unknown"
(or empty) produced event files that pinned the observer checkpoint and blocked
PIHA for ~34 days (see fix d5139c9). The agent must now refuse to start rather
than ever emit such a file.
"""
from __future__ import annotations
import os
from pathlib import Path
import pytest
from pydantic import ValidationError
from ha_diag import config
from ha_diag.config import Settings
@pytest.fixture(autouse=True)
def _isolate_environ():
"""Snapshot os.environ — Settings.load() mutates it via setdefault()."""
saved = dict(os.environ)
yield
os.environ.clear()
os.environ.update(saved)
# ---------------------------------------------------------------------------
# node_name loaded from env NODE_NAME
# ---------------------------------------------------------------------------
def test_node_name_from_env(monkeypatch):
monkeypatch.setenv("NODE_NAME", "piha")
settings = Settings()
assert settings.node_name == "piha"
def test_node_name_from_env_case_insensitive_var(monkeypatch):
# BaseSettings is case-insensitive: lowercase env var maps to the field too.
monkeypatch.delenv("NODE_NAME", raising=False)
monkeypatch.setenv("node_name", "chelsty-infra")
settings = Settings()
assert settings.node_name == "chelsty-infra"
def test_node_name_whitespace_is_stripped(monkeypatch):
monkeypatch.setenv("NODE_NAME", " piha ")
settings = Settings()
assert settings.node_name == "piha"
# ---------------------------------------------------------------------------
# fail-fast: unknown / empty node_name refuses to start
# ---------------------------------------------------------------------------
def test_explicit_unknown_node_name_refuses(monkeypatch):
monkeypatch.delenv("NODE_NAME", raising=False)
with pytest.raises(ValidationError) as exc:
Settings(node_name="unknown")
assert "refusing to start" in str(exc.value)
def test_unknown_case_insensitive_refuses(monkeypatch):
monkeypatch.delenv("NODE_NAME", raising=False)
with pytest.raises(ValidationError):
Settings(node_name="UNKNOWN")
def test_empty_node_name_refuses(monkeypatch):
monkeypatch.delenv("NODE_NAME", raising=False)
with pytest.raises(ValidationError):
Settings(node_name="")
def test_whitespace_only_node_name_refuses(monkeypatch):
monkeypatch.delenv("NODE_NAME", raising=False)
with pytest.raises(ValidationError):
Settings(node_name=" ")
def test_env_unknown_refuses(monkeypatch):
monkeypatch.setenv("NODE_NAME", "unknown")
with pytest.raises(ValidationError):
Settings()
def test_default_without_env_refuses(monkeypatch):
# No NODE_NAME anywhere → the "unknown" default must be rejected, not used.
monkeypatch.delenv("NODE_NAME", raising=False)
monkeypatch.delenv("node_name", raising=False)
with pytest.raises(ValidationError):
Settings()
# ---------------------------------------------------------------------------
# precedence: env wins over YAML for node_name
# ---------------------------------------------------------------------------
def test_env_wins_over_yaml(monkeypatch, tmp_path: Path):
yaml_file = tmp_path / "ha-diag-agent.yaml"
yaml_file.write_text("node_name: solaria\n")
monkeypatch.setattr(config, "_CONFIG_YAML", yaml_file)
monkeypatch.setenv("NODE_NAME", "piha")
settings = Settings.load()
# env value must win; YAML only supplies a default when env is absent.
assert settings.node_name == "piha"
def test_yaml_supplies_node_name_when_env_absent(monkeypatch, tmp_path: Path):
yaml_file = tmp_path / "ha-diag-agent.yaml"
yaml_file.write_text("node_name: chelsty-infra\n")
monkeypatch.setattr(config, "_CONFIG_YAML", yaml_file)
monkeypatch.delenv("NODE_NAME", raising=False)
monkeypatch.delenv("node_name", raising=False)
settings = Settings.load()
assert settings.node_name == "chelsty-infra"
def test_yaml_unknown_still_refuses(monkeypatch, tmp_path: Path):
# Even if YAML carries the poison value, fail-fast must still fire.
yaml_file = tmp_path / "ha-diag-agent.yaml"
yaml_file.write_text("node_name: unknown\n")
monkeypatch.setattr(config, "_CONFIG_YAML", yaml_file)
monkeypatch.delenv("NODE_NAME", raising=False)
monkeypatch.delenv("node_name", raising=False)
with pytest.raises(ValidationError):
Settings.load()
def test_load_valid_node_name(monkeypatch, tmp_path: Path):
monkeypatch.setattr(config, "_CONFIG_YAML", tmp_path / "does-not-exist.yaml")
monkeypatch.setenv("NODE_NAME", "piha")
settings = Settings.load()
assert settings.node_name == "piha"